Metabolic reprogramming occurs during tumorigenesis and holds promise for cancer therapy. However, the underlying mechanisms that control these metabolic alterations and their contribution to tumor maintenance are generally lacking. Mutations in NRF2 and its negative regulator KEAP1 are found in 15-34% of non-small cell lung cancer (NSCLC) and lead to constitutive NRF2 activity. Many NRF2-regulated processes center on the production and utilization of the antioxidant glutathione, which is synthesized from the amino acids cysteine, glycine and glutamate. In our previous work, we found that glutathione synthesis was critical for the pro-proliferative effects of NRF2 during tumor initiation. Furthermore, we found that NRF2 promotes metabolic rewiring to increase glycine availability to support glutathione synthesis. To define how constitutive NRF2 activity induces metabolic reprogramming to support GSH synthesis, we generated genetically engineered, conditional knock-in mouse models of the cancer mutations NRF2D29H and KEAP1R554Q. Our preliminary analysis of NRF2-regulated metabolism indicates that cysteine dioxygenase (CDO1) limits glutathione production by NRF2, produces toxic byproducts, and is silenced during tumorigenesis. We have established a metabolomics platform for the analysis of sulfur-containing metabolites, and a genetically engineered NRF2 and KEAP1 mutant lung cancer mouse models for in vivo studies and are now poised to define the tumor suppressive role of CDO1 and its metabolites during lung tumorigenesis. In Aim 1 we will examine the whether CDO1 antagonizes the NRF2-regulated antioxidant response by depleting cysteine. In Aim 2 we will examine the selective toxicity of CDO1 expression to cells with NRF2 activity due to toxic byproduct production. In Aim 3 we will examine whether CDO1 loss promotes lung tumorigenesis using our genetically engineered KEAP1 and NRF2 mutant mouse lung tumor models and patient tumor samples. The ultimate goal and the overall impact of this project are to characterize CDO1 as a novel metabolic liability for tumors with NRF2/KEAP1 mutations in order to define opportunities for therapeutic intervention for patients with these mutations, which currently lack targeted therapy.